q_rf_LUR: Quantile Random forest (using ranger, quantregForest)

In mengluchu/APMtools: statistical air pollution mapping tools

Description

Quantile Random forest (using ranger, quantregForest)

Usage

q_rf_LUR(
  variabledf,
  upq = "quantile= 0.05",
  lq = "quantile= 0.95",
  numtrees = 1000,
  mtry = NULL,
  y_varname = c("day_value", "night_value", "value_mean"),
  training,
  test,
  grepstring,
  ...
)

Arguments

variabledf	the dataframe containing predictors and dependent variable
upq	upper quantile default 0.95
lq	lower quantile, default 0.05
y_varname	name of the dependent variable.
training	the index for the rows used for training.
test	the index for the rows used for testing.
grepstring	the variable/column names of predictors in Lasso, grepl stlye, e.g. 'ROAD|pop|temp|wind|Rsp|OMI|eleva|coast'

Value

an error matrix and coverage probability, default 0.90 prediction interval.

mengluchu/APMtools documentation built on Jan. 27, 2022, 2:41 a.m.